Healthcare Analytics and Applications: Impact of Information Technology on Service Quality


Student thesis: Doctoral Thesis

View graph of relations


Related Research Unit(s)


Awarding Institution
Award date6 Jul 2020


The healthcare industry continues to face difficulty in improving care delivery efficiency, promoting care quality, and lowering clinical expenditures. Even though health information technology (HIT) has a widely acknowledged role in addressing these concerns, there is mixed and inconclusive evidence regarding HIT impact in the existing literature, which requires further investigation. In this dissertation, we develop three studies to explore how HIT can transform healthcare delivery by addressing the workload issue, improving in-hospital service quality, and alleviating medical resource overuse.

In the first study, we focus on the workload issue since it poses a significant challenge to the delivery of high-quality healthcare services. To this end, several studies have investigated the impact of workload on healthcare quality. However, how workload in different core tasks affects care quality remains an open question. Furthermore, while Electronic Medical Record (EMR) enables improvement of work processes, there is no evidence on how EMR use interacts with physician workload under different tasks. We believe this is a research gap surrounding the interaction effects of workload on task and quality under information technology in healthcare, or simply “interaction effects of workload in healthcare.” We draw on the operation management literature and the task-technology fit (TTF) theory and use an operational-level panel dataset from a major hospital to examine various vital issues on interaction effects in terms of patient readmission and medical errors. Our empirical results show that care quality can be nonlinearly affected by diagnostic and treatment workload upon physicians. In addition, given different EMR features, the results uncover that advanced EMR use can help physicians alleviate the negative effect of variant workload on care quality under various tasks, and such effect is more substantial at the lower and higher ends of workload. In contrast, we find that basic EMR use improves care quality only under a lower level of diagnostic and treatment workload. Furthermore, continuous usage of basic EMRs at the high end of workload turns to enlarge the negative workload effect on care quality. Our findings in this study not only establish new quantitative insights surrounding the interaction effects of workload but also extend the utility of TTF towards complex tasks with high business value, particularly in the context of healthcare.

In the second study, we investigate how physicians provide online healthcare services through health social media (HSM) to affect the performance of their offline services in hospitals. Although online healthcare services have the potential to transform care delivery processes by reaching out directly to patients, existing literature has not investigated how online services would affect offline services in the healthcare context. Given this research gap, we propose two research questions: Can physicians’ use of HSM improve their in-hospital service quality? Moreover, if so, what are the underlying mechanisms behind? To study these questions, we assemble a novel dataset that combines physicians’ HSM adoption with the 236,557 patient records of visits to a large hospital in China. We analyze the data by using difference-in-difference analysis, propensity score matching method, and instrumental variable approach. By drawing upon process virtualization theory, our results show that physicians’ HSM use is negatively associated with length of stay and thirty-day readmission. In addition, the magnitude of such effects is increased when patients have restricted channels to communicate with physicians, and they have limited health literacy. Collectively, our results reveal that HSM use can improve care quality through gatekeeper removing and knowledge sharing mechanisms. This study not only pushes the research frontier on the value of HSM on the online-to-offline service model in the healthcare context but also inspires physicians on how to leverage HSM to improve in-hospital service quality through the mechanisms we discover.

In the third study, we further investigate the impact of HSM on the overuse of medical resources (OMR), which is a significant contributor to the rapidly growing healthcare costs. Although there is a large body of literature on healthcare values of information technology, very few studies examine the impact of health IT in this regard. Besides, in comparison to traditional clinical HITs, the impact of patient-centric HIT, and how it interacts with clinical HITs have been less investigated. Motivated by these crucial research gaps, we collect a data set consisting of 33,526 diabetes patients from a large hospital in China. Guided with media richness theory, we apply an instrumental variable approach to examine the impact of HSM on high-cost imaging tests (HCIT), which can be overused in many cases. Our results indicate that HSM can reduce the use of HCIT without lowering care quality. In addition, the interplay between HSM and CPOE can further enhance this effect. Our research shows the significant value of IT application in resolving a social problem. In addition, we uncover the importance of health IT complementarity by integrating patient-centric systems with clinical health ITs.